A computional model for heart disease prediction by using data mining approach

Serval diseases can be diagnosed using machine learning techniques, but the focus of this work will be on heart disease diagnosis.   For our work, we are using residual neural network (ResNet) for feature extraction and Recurrent Neural Network (RNN) for training. ResNet is a type

Project Title

A computional model for heart disease prediction by using data mining approach

Project Area of Specialization

Software Engineering

Project Summary

Serval diseases can be diagnosed using machine learning techniques, but the focus of this work will be on heart disease diagnosis.

For our work, we are using residual neural network (ResNet) for feature extraction and Recurrent Neural Network (RNN) for training. ResNet is a type of specialized neural network that helps to handle more sophisticated deep learning tasks and models. It deals with problems like accuracy saturation, degradation, vanishing gradients and curse of dimensionality by using residual blocks, which take advantage of residual mapping to preserve inputs. RNN architectures is that error gradients vanish exponentially quickly with the size of the time lag between important events and favorable for classification prediction, regression prediction problems and generative models.

This work provides an overview of the machine learning classification techniques used in the field of diagnosing heart disease, and how previous researchers implemented them. It throws the light on how important is machine learning in the healthcare field and how it can make accurate predictions and help healthcare professionals.

Project Objectives

The main objective of this research is to develop a prototype Intelligent Heart Disease Prediction System with CANFIS and genetic algorithm. Cleveland Dataset with 14 attributes is used. The objective of this project is to develop a machine learning based model for efficient heart disease prediction. It is difficult to manually determine the odds of getting heart disease based on risk factors. So using machine learning techniques our model would be able to predict the output from existing data.

In the end efficient model for the heart disease prediction with improved accuracy is expected. Since heart disease is the primary cause of deaths in the world today, and the effective diagnosis of heart disease is immensely useful to save lives. However, 90% of those deaths were estimated to be preventable if patients have correctly been diagnosed early and they improved their habits such as: healthy eating, exercise, and alike.

Project Implementation Method

The purpose of this study is to analyze different data mining techniques proposed in recent years for prediction of heart disease. The paper mentions five techniques of knowledge gathering via data mining techniques. The first one is analysis of dataset via Tangra Tool using 3 supervised Machine Learning algorithms i.e. Naive Bayes, KNN and decision list algorithm. The Naïve Bayes algorithm provides better results with 52.33% accuracy in prediction of heart disease in least time of 609ms. The second technique uses .NET Platform to implement data mining techniques i.e. Naïve Bayes, Decision Trees and Neural Networks. The Naïve Bayes gives highest correct prediction of heart disease with percentage of 86.53, followed by ANN with predictive percentage of 85.53. Another Technique involves the use of Genetic Algorithm. To diagnose the presence of heart disease three classifiers of machine learning are used in it i.e. Decision Tree, Naïve Bayes and classification through clustering. The decision tree has accuracy of 99.2% followed by consistent results of Naïve Baye’s with 95.6%. Clustering lacked the accuracy of classification with percentage of 88.3. Further two techniques are Association Rule Discovery and Rough Set Theory. The results obtained via RST shows that it can select small number of rules from a large number of rules without any probable change in quality of classification. [2]

Data Mining Classification techniques are used in this paper for diagnosis of heart disease. The main objective of this research is to build Intelligent Heart Disease Prediction System that gives diagnosis of heart disease using historical heart database. In this Research Papers, different data mining classification techniques have been applied for diagnosis of heart disease like viz. Neural Networks, Decision Trees, and Naive Bayes. As a result, Naïve Bayes’ Accuracy with 13 attributes is 94.44% and with 15 attributes is 90.74%. Decision trees’ accuracy with 13 attributes is 96.66% and with 15 attributes is 99.62%. Neural Networks’ accuracy with 13 attributes is 99.25% and with 15 attributes is 100%.  Therefore, three data mining classification techniques were applied namely Decision trees, Naive Bayes & Neural Networks. From results it has been seen that Neural Networks provides accurate results as compare to Decision trees & Naive Bayes.

In this research, an advanced deep neural network approach is used to predict coronary heart disease in patients and increase diagnostic accuracy using classification and prediction models.

Benefits of the Project

The objective of this project is to develop a machine learning based model for efficient heart disease prediction. It is difficult to manually determine the odds of getting heart disease based on risk factors. So using machine learning techniques our model would be able to predict the output from existing data.

Heart disease diagnosis is one of the most critical and challenging tasks in the healthcare field. It must be diagnosed quickly, efficiently and correctly in order to save lives. It requires the patient to do many tests, and healthcare professionals must carefully examine the results. That is why developing different heart disease prediction systems using various machine learning algorithms will provide accurate result in less time.

Following are some of the other benifits:

• the system can predict heart diseases
• It is for the health measurement
• It will decrease the death rate due to heart disease
• It will be helpful to the society
• It will be helpful for doctors and patients
• It will be helpful for medical students
• It will also be good for medically uneducated people
• It will be useful for the society

Technical Details of Final Deliverable

Four databases are generally used Cleveland, Hungary, Switzerland, and the VA Long Beach. These databases contain 76 attributes, but all published experiments refer to using a subset of 14 of them. In particular, the Cleveland database is the only one that has been used by ML researchers to this date. It is integer valued from 0 (no presence) to 4. Experiments with the Cleveland database have concentrated on simply attempting to distinguish presence values 1, 2, 3, 4) from absence (value 0). 

Generic ML techniques will be used for this data to predict heart disease, which will further be refined by ensemble of ML techniques.

The health care industry produces a huge amount of data. Using this huge amount of data, a disease can be detected, predicted or even cured. But unfortunately, this data is not always made use to the full extent and is often underutilized.

A huge threat to human kind is caused by diseases like heart disease, cancer, tumor and Alzheimer’s disease. Heart related diseases or Cardiovascular Diseases (CVDs) are the main reason for a huge number of death in the world over the last few decades and has emerged as the most life-threatening disease, not only in developing countries but throughout the globe. Therefore, there is a need of reliable, accurate and feasible system to diagnose heart diseases in time for proper treatment. 

In recent times, Machine Learning algorithms and techniques have been applied to various medical datasets to automate the analysis of large and complex data to help the health care industry and the professionals in the diagnosis of heart related diseases. Many researchers became interested in using machine learning for diagnosing diseases because it helps to reduce diagnosing time and increases the accuracy and efficiency.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Health

Other Industries

Education , IT , Medical

Core Technology

Wearables and Implantables

Other Technologies

Artificial Intelligence(AI)

Sustainable Development Goals

Good Health and Well-Being for People

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com